product · May 17, 2026
Microsoft Adds DeepSeek V4 Pro and Kimi K2.6 to Fireworks AI on Foundry
Share the canonical public link.
Microsoft expanded Fireworks AI offerings on Foundry on May 14, 2026 with DeepSeek V4 Pro and Kimi K2.6. Fireworks AI inference layer supports low-latency high-throughput workloads natively in Azure. Kimi K2.6 serverless pricing is $0.95 per million input tokens, $0.16 per million cached tokens and $4.00 per million output tokens. DeepSeek V4 Pro serverless pricing is $1.75 per million input tokens, $0.15 per million cached tokens and $3.48 per million output tokens. US Data Zone PTUs became generally available for data sovereignty in regulated industries.
Spend governor blocked model creation: provider_circuit_open (lane=dev, provider=together)